Haptic Stiffness Perception Using Hand Exoskeletons in Tactile Robotic Telemanipulation
Gabriele Giudici, Claudio Coppola, Kaspar Althoefer, Ildar Farkhatdinov, Lorenzo Jamone
AI summary
Problem
While visual feedback dominates teleoperation, haptic feedback for perceiving object properties like stiffness remains underexplored, particularly for human operators using hand exoskeletons without visual cues.
Approach
Ten participants teleoperated a robotic hand via a hand exoskeleton to squeeze soft objects, comparing two haptic feedback methods: force-only feedback versus force combined with kinematic displacement mismatch feedback.
Key results
- 75% average accuracy in identifying the most similar object to a reference
- 65% average accuracy in determining which object in a pair was softer
- First demonstration of stiffness discrimination using only haptic feedback in telemanipulation
- Kinematic displacement feedback improved discrimination for objects with similar stiffness
Why it matters
Enables safer and more effective teleoperation in healthcare and hazardous environments where visual feedback is unavailable or insufficient for assessing material properties.
Abstract
Robotic telemanipulation is central in many appli- cations, from healthcare to harsh environments. While visual feedback from cameras can provide valuable information to the human operator, haptic feedback offers insight into certain object properties - such as stiffness - that vision alone cannot provide. However, the use of haptic feedback alone has been largely unexplored. To bridge this gap, we tested ten participants to remotely squeeze soft objects to perceive their stiffness, by teleoperating a dexterous robotic hand using an active hand exoskeleton. Two haptic feedback methods were compared: using only the contact forces measured with the tactile fingertips of the robotic hand, or including a kinematic measure as well (the motion mismatch between the hand exoskeleton and the robotic hand). Our results demonstrate, for the first time, that operators using a hand exoskeleton are indeed capable of discriminating objects of different stiffness relying on haptic feedback alone, with an average accuracy of 75% to identify which object in a pair was most similar to a reference, and 65% to determine which object in a pair was softer. In addition, our findings also suggest that including a kinematic measure in the feedback may enhance discrimination between objects of similar stiffness.